为解决传统电网故障分类器无法准确获得故障线路的信息及定位故障的问题,本文提出了基于深度学习的电网故障预警系统。网络中增加由BIGRU提取的故障线路时间序列特征,从而提高分类器的精度。此外,通过注意力机制学习不同的故障线路或不同的故障状态的时间序列特征,从而加快网络学习效率。最后,将本文所提框架与LSTM、GRU、BILSTM、BIGRU等网络进行对比,本文所提方法预测准确率可达92.9%。

In order to solve the problem that the traditional power grid fault classifier can not accurately obtain the fault line information and locate the fault, a power grid fault early warning system based on deep learning is proposed.The fault line time series features extracted by BI-GRU are added to the network to improve the accuracy of the classifier.In addition, the attention mechanism is used to learn the time series characteristics of different fault lines or different fault states, so as to speed up the network learning efficiency.Finally, the proposed framework is compared with LSTM, GRU, BILSTM, BiGRU and other networks, and the prediction accuracy of the proposed method can reach 92.9%.